The SpaceX Thread

[UserJoe]

That depended on having propulsive landing for Dragons, because Mars doesn't have oceans or much atmosphere. They thought they could develop propulsive landing for Dragons the same way they did for Falcon 9, by using paid-for customer flights. Specifically, cargo return trips from ISS, paid for by NASA. What changed is that NASA (quite reasonably) refused to let SpaceX risk their down-cargo. For SpaceX to test propulsive Dragons on their own dime would have been expensive. So it got cancelled, in favour of going all-in on Starship. One of SpaceX virtues is their willingness to change their plans when necessary.

Starship landing on Mars in 2027 was plausible when the first propellant transfer tests were happening 2Q 2025, which itself was plausible in 4Q 2024. Now the setbacks with V2 Starship are making it look much less likely. The orbital propellant transfer tests may not happen until next year now, and of course they need to land Starship on Earth before attempting it on Mars. I still wouldn't rule it out. They'll surely pull out all the stops to avoid losing the 2-year transit window.

They probably need to pull out all the stops to meet their NASA commitments for the moon landing system at this point. Their engineering method of measure once, cut n times, solve for n seems to have run into problems with Starship.

 

[deadman12-4]

That depended on having propulsive landing for Dragons, because Mars doesn't have oceans or much atmosphere. They thought they could develop propulsive landing for Dragons the same way they did for Falcon 9, by using paid-for customer flights. Specifically, cargo return trips from ISS, paid for by NASA. What changed is that NASA (quite reasonably) refused to let SpaceX risk their down-cargo. For SpaceX to test propulsive Dragons on their own dime would have been expensive. So it got cancelled, in favour of going all-in on Starship. One of SpaceX virtues is their willingness to change their plans when necessary.

Starship landing on Mars in 2027 was plausible when the first propellant transfer tests were happening 2Q 2025, which itself was plausible in 4Q 2024. Now the setbacks with V2 Starship are making it look much less likely. The orbital propellant transfer tests may not happen until next year now, and of course they need to land Starship on Earth before attempting it on Mars. I still wouldn't rule it out. They'll surely pull out all the stops to avoid losing the 2-year transit window.

Starship landing on mars was NEVER plausible in that timeline. Its been just 2-3 years away since 2018. Almost like fusion - just around the corner but the corner never actually gets closer.
Starship landing on mars is much further away, but the story is the real goal. Making people believe something that isn't true.

 

[Dmytry]

They probably need to pull out all the stops to meet their NASA commitments for the moon landing system at this point. Their engineering method of measure once, cut n times, solve for n seems to have run into problems with Starship.

To be honest the Starship is just a huge complex rocket, more like the Soviet N1 than Saturn V.

The reason N1 kept exploding was that the most economical way to fully test them was to launch it. Again and again.

Of course, the propagandist spin on this is that soviets simply weren't good at engineering a rocket, but I think that (apart from some significant and yet easily overstated advantage from CAD software and simulation), the engineering is and was about as good as it could realistically be.

As far as costs go the accounting was pretty funky under communism, and it is even more funky under modern capitalism as well - the "broken window fallacy" is partially true when there's a huge net flow of wealth towards its concentration, making even wasteful non-concentration activities (like paying the engineers) better than the alternative (where instead of making a rocket, engineers are employed in less productive roles, perhaps as unproductive as making videogames).

 

[deadman12-4]

To be honest the Starship is just a huge complex rocket, more like the Soviet N1 than Saturn V.

The reason N1 kept exploding was that the most economical way to fully test them was to launch it. Again and again.

Of course, the propagandist spin on this is that soviets simply weren't good at engineering a rocket, but I think that (apart from some significant and yet easily overstated advantage from CAD software and simulation), the engineering is and was about as good as it could realistically be.

Talk about alternate history. The N1 wasn't being "tested", they were honestly trying to make it work. They were not eschewing computer simulations to work everything out and were simply trying to be "hardware rich" to "move fast and break things". This is a fantasy rewrite of history.

 

[Dmytry]

Talk about alternate history. The N1 wasn't being "tested", they were honestly trying to make it work. They were not eschewing computer simulations to work everything out and were simply trying to be "hardware rich" to "move fast and break things". This is a fantasy rewrite of history.

I have trouble parsing what is it that you are trying to say. That SpaceX eschews computer simulations when it comes to starship? That USSR engineers did?

N1 was spectacularly blowing up, again and again, because even with all the test firings they could do at a test stand, a number of integration issues which would inevitably slip manual review, had to get discovered in flight. Then it was cancelled because it had one purpose, to go to the moon, and that became irrelevant.

Likewise, starship is a huge huge rocket with approximately the same number of engines. Plus new shit like re-lighting said engines.

The similar reason for cancellation is unlikely to occur, but another one could - it is all done on a whim of a billionaire, who has plenty of other whims.

 

[Megalodon]

N1 was spectacularly blowing up, again and again, because even with all the test firings they could do at a test stand, a number of integration issues which would inevitably slip manual review, had to get discovered in flight.

Point of clarification: they couldn't do a static test of the N1 because the NK-15 engines used could only be fired once. The subsequent NK-33 engine resolved this issue but was never flown as part of the Soviet program because as you note the program was wound down (though it did fly on an American launcher decades later). Subsequent Soviet launchers like Zenit and Energia ultimately used the oxygen rich staged combustion technology, though not using the NK-33 engine. Fortunately the later RD-170 engine family shook the bad reputation, and are some of the most remarkable engines ever made.

 

[Dmytry]

Ahh, right. I mean, the engine design was obviously developed via stand testing, but the integration they had to test by flying it, expecting a number of failures.